Content uploaded by Vinita Agarwal
Author content
All content in this area was uploaded by Vinita Agarwal on Aug 12, 2016
Content may be subject to copyright.
Investigating the convergent
validity of organizational trust
Vinita Agarwal
The Department of Communication Arts,
Salisbury University, Salisbury, Maryland, USA
Abstract
Purpose – Organizational trust is an important construct for organizational and public relations
scholars and practitioners for its influence on key organizational outcomes, yet the convergent validity
of organizational trust instruments has not been investigated by any study. The purpose of this paper
is to address an important gap in the literature by reporting the results of a systematic investigation of
the convergent validity of three organizational trust measures, taking an interpersonal, public
relations, and organizational approach to trust in organizations respectively.
Design/methodology/approach – IRB approval was obtained for a cross-sectional study design
gathering self-reports from participants through an online data gathering system of a large
Midwestern university in the USA. Correlational matrices, along with exploratory and confirmatory
factor analyses adapting the multitrait-multimethod matrix, were employed for data analysis.
Findings – The three trust measures demonstrate mixed evidence of convergent validity with each
other and with a theoretically correlated construct, organizational identification, demonstrating that
while the three measures share the same conceptual ground, there is a need to clarify their underlying
theoretical specification, especially with respect to organizational identification.
Research limitations/implications – Future large-scale studies can extend the findings based on a
student population by employing multiple and diverse data sets, as well as investigate method variance.
Practical implications – Recommendations to improve convergent validity include: increasing item
parsimony to decrease redundancy; revise item construction; and improved theoretical specification
investigating the conceptual boundaries organizational identification with trust.
Originality/value – The study provides empirical evidence of the need for greater conceptual and
operational clarification of the theoretical bases of trust. It is the first to advance research on trust in
organizations by providing empirical support for the convergent validity of organizational trust
measures assessed from organizational, interpersonal, and public relations perspectives.
Keywords Organizations, Trust, Construct validity, Convergent validity, Discriminant validity,
Organizational trust, Organizational identification
Paper type Research paper
Organizational trust is of increasing interest to scholars for its positive influence on
organizational and member outcomes including organizational effectiveness, job
satisfaction, and team performance (Argyris, 1962; Coleman, 1990; Rousseau et al.,
1998; Tannenbaum et al., 1992). Recent scholarly interest in trust has led to a diversity
of theoretical perspectives in trust research (McAllister, 1995). For example, trust
research can be categorized by level of analysis ranging from micro-level interpersonal
approaches examining individual dispositions (Rotter, 1967, 1971, 1980), socio-psychological
approaches examining environmental factors (Bhattacharya et al., 1998; Lewicki and
Bunker, 1995), to macro-level institutional approaches examining uncertainty in societal
interactions (Zucker, 1986) and trust. Trust research can also be categorized as antecedent-
based research and outcome-based research. For example, researchers examining the
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1363-254X.htm
Received 28 July 2010
Revised 11 January 2011
10 May 2011
Accepted 15 July 2011
Journal of Communication
Management
Vol. 17 No. 1, 2013
pp. 24-39
rEmerald Group Publishing Limited
1363-254X
DOI 10.1108/13632541311300133
The author would like to thank the two anonymous reviewers for their helpful comments on
earlier drafts of this article.
24
JCOM
17,1
antecedents of trust-based beliefs may take a calculus-based approach, knowledge-based
approach, or an identification-based approach to explain the development of the feelings,
attitudes, and beliefs that characterize it ( Jones and George, 1998).
Measures of trust in organizational scholarship evaluate trust in a variety of
organizational contexts such as dyadic manager-employee interaction (organizational
trust inventory, OTI, Cummings and Bromiley, 1996) or trust in virtual teams
( Jarvenpaa et al., 1998). However, despite the development of measures of trust in
the literature, the construct validity of key instruments employed to measure
organizational trust has not been assessed by any study. Assessing the validity and
reliability of measurements of organizational trust is important, first, because it
confirms that the instrument is theoretically grounded and its operationalization
is credible for the interpretation of research goals and findings. Second, invalid
measurements can confound the goals of future research and interpretations which in
turn, obscures our understanding of the trust in organizations. Third, investigating the
construct validity of organizational trust is important because much of current trust
research has employed variable-based self-reports to investigate trust in organizational
relationships. In essence, establishing the construct validity of a measure involves
demonstrating the convergent validity and the discriminant validity of the measure.
The present study addresses an important gap in the literature by reporting the results
of a systematic investigation of the convergent validity of three organizational trust
measures taking an interpersonal, public relations, and organizational approach to
trust in organizations, respectively. These measures were: Cook and Wall’s (1980)
interpersonal trust in management; Hon and Grunig’s (1999) organizational trust; and
Ellis and Shockley-Zalabak’s (2001) trust in top management.
This paper is organized as follows. The first section provides a theoretical overview
of the construct of organizational trust and establishes the dimensions of the
theoretical relationships operationalized by the three instruments. Following this, in
the second section, the construct of organizational identification and its relationship
with organizational trust is explicated. The third section explains the theoretical bases
of construct validity and the methods employed to investigate convergent validity.
Fourth, and finally, the hypotheses tested in the study are laid out followed by the
methods, analysis, results, and discussion sections.
1. Trust in organizational contexts
Trust in organizations is distinct in nature from other contexts (e.g. trust in dating
relationships, Larzelere and Huston, 1980). Organizational trust has been approached
from a variety of perspectives. For example, from a psychological perspective, Rotter
(1967) defines trust as “a generalized expectancy held by an individual or group that
the word, promise, verbal, or written statement of another individual or group can be
relied upon” (p. 444). Social psychologists take a state-based view of trust as “the
willingness of a party to be vulnerable to the actions of another party based on
the expectation that the other party will perform a particular action important to the
trustor, irrespective of the ability to monitor or control the party” (Mayer et al., 1995,
p. 712). Economists (such as Dasgupta, 1998) adopt a calculus-based cost-benefit
approach to trust as regulating fair behavior. Bhattacharya et al. (1998) integrate
micro- and macro-approaches and define trust as “an expectancy of positive
(or nonnegative) outcomes that one can receive based on the expected action of another
party in an interaction characterized by uncertainty” (p. 462). A common theme
underlying all dominant approaches in trust research is an agreement on the
25
Organizational
trust
antecedent conditions necessary for trust-based behavioral orientations to arise.
These are: a risk-based context characterized by uncertainty and vulnerability, an
interdependence between two parties in that communicative context, and, a shared
goal (Bhattacharya et al., 1998; Lewicki and Bunker, 1995).
A state-based definition of trust in organizations is distinct from an interpersonal
view of trust (Rotter, 1967) and from an economical, transaction-based trust
(Das and Teng, 1998) in two key ways. First, a state-based definition of organizational
trust specifies the boundary condition as the relationship between the trustor and
the trustee; and second, it examines the influence of the organizational context on
the development of trust (Mayer et al., 1995). Rousseau et al. (1998) provide a
cross-disciplinary definition of state-based trust as a “psychological state comprising
the intention to accept vulnerability based upon positive expectations of the intentions
or behavior of another” (p. 395) integrates most trust-based theorizing in
organizational contexts (e.g. Ellis and Shockley-Zalabak, 2001; Mayer et al., 1995).
The sub-sections below focus on organizational trust and review the three main
approaches taken by researchers in organizational contexts: interpersonal trust in top
management, trust in top management as a relationship outcome from public relations
perspective, and trust in top management from organizational communication perspective.
1.1 Interpersonal trust in top management
Interpersonal approaches examine trust as a belief or an attitude that is influenced by
the characteristics of both the communicative interactions and the context (Rotter ,
1967, 1980; Whitener et al., 1998). Interpersonal trust is categorized along two
dimensions – cognitive dimension (perception of integrity or capability of the trustor)
and affective dimension (perception of concern, care, or benevolence of the trustor
toward the trustee, McAllister, 1995). Cook and Wall (1980) take an interpersonal
approach to conceptualize individual-level employee trust in top management.
In their measure, Cook and Wall (1980) examine trust at work to measure trust as a
“directly experienced evaluative or affective reaction” (p. 40) between the trustor and
the trustee. Their study provides evidence for the stability of its psychometric
properties in blue collar UK-based employees (Cook and Wall, 1980). The two main
dimensions in Cook and Wall’s (1980) scale are: faith in trustworthy intentions of the
trustor, and confidence in the ability of the trustor. The study also finds that employee
trust in management (faith in management and confidence in management) is strongly
associated with organizational commitment.
1.2 Organizational trust in public relations
On the other hand, trust in public relations has been conceptualized as an outcome-
based variable characterizing the relationship between two key organizational publics:
organizational members and management. Hon and Grunig’s (1999) measure
of organizational trust is conceptually grounded in the public relations theoretical
perspective. It conceptualizes organizational members as strategic publics in a
two-way relationship with the management toward the goal of achieving mutually
beneficial individual and organizational outcomes. Trust in the organization is one of
the six relationship outcome measures (including control mutuality, satisfaction,
commitment, exchange relationship, and communal relationship) proposed by Hon
and Grunig (1999) to measure the strength of the relationship between organizations
and their publics. The organizational trust scale employed by Hon and Grunig
(1999) consists of three constructs similar to the dimensions employed by
26
JCOM
17,1
Ellis and Shockley-Zalabak (2001) and to the managerial trustworthy behaviors (MTB)
characteristics employed by Cook and Wall (1980) because it also identifies the
key behavioral characteristics of the organization that lead to the development of
evaluative trust-based beliefs in the publics.
The three dimensions of Hon and Grunig’s (1999) scale are: integrity, the belief that
the organization is fair and just; (b) dependability, the belief that an organization will do
what it says it will do; and (c) competence, the belief that an organization has the ability
to do what it says it will do. Respondents chose a number from 1 to 9 to indicate
the extent to which they agree that the item described their relationship with the
organization. The reliabilities for the 11-item trust scale reported in the original study
were: a¼0.90 (GE), 0.89 (NRA), 0.93 (social security), 0.91 (Microsoft), 0.91 (Red Cross)
(with an average a¼0.91).
1.3 Trust in top management in organizational communication
Ellis and Shockley-Zalabak (2001) take the approach that organizational trust is
“communication-based, dynamic, multi-faceted” (p. 383). Ellis and Shockley-Zalabak
(2001) define trust in top management or organization as “positive expectations
about the behavior of others based on roles, relationships, experiences, and
interdependencies” (p. 383). Theoretically, the development of the Ellis and
Shockley-Zalabak’s (2001) scale was guided by Mishra’s (1996) model identifying the
four dimensions of trust as: competence in organizational leadership and organization
(Gabarro, 1987); openness, honesty, and sincerity in communication (Whitener et al.,
1998); leadership caring and concern of organizational members (Cummings and
Bromiley, 1996); and reliability or an expectation of consistency, dependability, and
congruency between word and action of behavior in leadership (McGregor, 1967).
The trust in top management scale developed by Ellis and Shockley-Zalabak (2001) lays
greater emphasis on the role of communication in the “development and maintenance
of trust in top management” (Ellis and Shockley-Zalabak, 2001, p. 385) and was
developed through focus groups held in the USA and Italy utilizing the critical incident
technique. The main dimensions of trust extracted included: consistency in actions,
following through with stated intentions, dependability/reliability, sincerity, keeping
confidences, avoiding gossip, using direct eye contact, avoiding retaliation, standing up
for others, and engaging in open, honest discussion. The confirmatory factor analysis
(CFA) reported by the authors in the study demonstrates excellent model fit
(w
2
(8) ¼7.51, NNFI ¼0.99, CFI ¼0.99, GFI ¼0.99). Finally, the study reports a high
internal reliability ( ¼0.94, M¼2.74, SD ¼1.04) (Ellis and Shockley-Zalabak, 2001).
Although the three measures of trust in organizational contexts have taken different
approaches in constructing their operationalizations, there are several important ways
in which they are similar. For instance, Mishra’s (1996) model conceptualizing
trustworthiness that Ellis and Shockley-Zalabak (2001) draw upon for constructing
their measure is parallel in key ways to Cook and Wall’s (1980) conceptualization.
To illustrate, competence in Mishra’s (1996) model and ability in Cook and Wall’s (1980)
model share the same conceptual ground. Similarly, the factor of caring and
benevolence in interpersonal trust (McAllister, 1995) is similar to that of concern
(Mishra, 1996) and the factor of integrity in trustee behavior (Hon and Grunig, 1999)
is related to reliability as adopted by Mishra (1996) and later by Ellis and
Shockley-Zalabak (2001). The model of dyadic trust in organizational contexts adopted
by Cook and Wall (1980) is also similar to that followed by Ellis and Shockley-Zalabak
(2001) in examining the employee perception of the intentions and confidence of the
27
Organizational
trust
behaviors (words and actions) of the management. Thus while each measure is
based on different approaches, in several key ways the structure of Ellis and
Shockley-Zalabak’s (2001) dimensions is conceptually similar to the dimensions
employed by Hon and Grunig (1999), and those by Cook and Wall (1980). We should,
therefore, expect to see a high degree of positive correlation between the three
measures of organizational trust. The next section discusses the construct of
organizational identification and explicates its relationship with organizational trust.
2. Organizational identification
Organizational identification is defined as the process by which individuals link
themselves to the targets of their identification (Cheney, 1983), in this case,
organizations. It has been argued that organizations constitute a natural target for the
identification process of individuals as they fulfill a fundamental need to identify in
human relations (Burke, 1950; Cheney, 1983). The process of member identification has
been found to be closely related to organizational activities such as socialization,
communication, and personnel selection (Lee, 1971; March and Simon, 1958). From
a critical organizational perspective, the relationship building process of identification
also fulfills the function of communicating and managing the value premises and
decisional premises of the organization (Bullis and Bach, 1991; Cheney, 1983; Cheney
and Tompkins, 1987; Tompkins and Cheney, 1983). In their original study providing
a measure of organizational identification, Cheney (1983) reported an excellent reliability
of a¼0.95 and a single factor solution accounting for 86 percent of the variance.
Trust in organizations and organizational identification both imply alignment with
key organizational premises. However, inherent in the conceptual definition of trust is
an assumption that trust is more than just an attitude or belief or expectation of the
trustor toward the trustee, such that it also exists in the relationship between the two
(Whitener et al., 1998). Shockley-Zalabak et al. (2001) note that when organizational
trust is approached as an “aggregate across employees, partnerships, stockholders,
customers, and other stakeholders, [the] generalized expectations about intent and
behaviors become part of the cultural context of the organization” (p. 37). Other
studies provide indirect support to the hypothesis that organizational trust should
be positively associated with organizational identification. For example, studies
examining the relationship between MTBs and organizational factors find
organizational attributes such as culture and structure, where culture is defined as
“the set of shared, taken-for-granted implicit assumptions that a group holds” (Schein,
1996, p. 236) also influence the perception of MTBs (Creed and Miles, 1996; Hardin,
2000; Rousseau, 1990). Viewed in this manner, member identification with the cultural
values, norms, and decisional premises of the organization and should be strongly
associated with the trustworthy perceptions of that target.
Organizational identification has also been closely associated with organizational
trust. Researchers examining the antecedents of trust note that organizational trust
is composed of three dimensions based on the relative strength of the premises
characterizing its emergence: exchange-based trust (economical, cost-benefit analysis),
knowledge-based trust (on the predictability or confidence in another’s behavioral
orientations), or identification-based trust (through internalization of the other’s
motives and intentions). In its identification-based dimension, organizational trust
arises between members and the organization through the act of identification with or
internalization of the organization’s motives, values, or decisional premises, suggesting
again that organizational trust and organizational identification should be strongly
28
JCOM
17,1
correlated. While trust research has yet to establish the mechanisms through
which knowledge-based, exchange-based, and identification-based trust related to
other organizational processes such as organizational identification, Ellis and
Shockley-Zalabak (2001) find that organizational identification of members should
be strongly associated with organizational trust (Ellis and Shockley-Zalabak, 2001).
They suggest that organizational identification could comprise a dimension of
organizational trust. The present study extends this line of research by empirically
testing the hypothesis that organizational trust should be positively and strongly
associated with organizational identification.
3. Theoretical bases of construct validity
Theoretically, a construct (such as organizational trust), is defined as “a postulated
attribute of people assumed to be reflected in test performance” (Cronbach and Meehl,
1955, p. 283). In this regard, a construct (as an attribute or trait such as trusting
intentions) carries associated meanings that describe its observable properties.
The logic of construct validation involves reaching informed interpretations of its test
to validate these accurately reflect the meaning underlying the construct. Thus a test
of construct validity of organizational trust should provide a method for assessing
the theoretically defined meanings of the construct are measured through its
operationalization. In general, construct validation includes strategies estimating
the internal process, structure, or state to determine the validity and reliability of
the measurements. Essential to the construct validation process is defining the
interlocking net of laws known as the nomological network (Cronbach and Meehl,
1955). These laws relate: “(a) observable properties of the construct, (b) theoretical
constructs to observables, or (c) different theoretical constructs to one another”
(Cronbach and Meehl, 1955, p. 290). If a construct is to be scientifically admissible, it
should occur in a nomological net of testable theoretically based relationships.
Construct validity of trust is established when the researcher can generalize what is
observed to the underlying theoretical concept of the measures. Validation procedures
for establishing construct validity are theoretically based and include establishing
the convergent validity and discriminant validity of the measure. Overall, procedures
to establish construct validity include: tests that the correlational matrices between the
three organizational trust measures are associated as expected, examining the internal
factor structure of the organizational trust measures for interpretability, and tests
confirming the theoretically based relationship between organizational trust and
organizational identification is obtained as expected to demonstrate the convergent
validity of the measures (Cronbach and Meehl, 1955):
H1. The trust measure (Hon and Grunig, 1999) will demonstrate a positive relationship
between trust scores with trust in top management (Ellis and Shockley-Zalabak,
2001) and interpersonal trust in top management (Cook and Wall, 1980).
H2a. The organizational trust measures will show similar and comparative positive
and significant association with Cheney’s (1983) organizational identification.
H2b. The internal validity score of organizational trust (Hon and Grunig, 1999)
measure will be similar to the internal validity score of Ellis and
Shockley-Zalabak (2001) and to the interpersonal trust in top management
(Cook and Wall, 1980).
29
Organizational
trust
4. Method
The study participants (n¼109) were undergraduate students (97.2 percent of study
participants reported their age as between 18 and 25 years, 81 percent reported
ethnicity as Caucasian, 48 percent worked for small business, and 47 percent of
participants reported working between 6 months and 2 years at the time of taking the
survey). Study participants had the option to voluntarily sign up for participation after
reading the study description for extra credit at a large Midwestern university. In the
study description potential participants were informed that they would be asked about
“trust and commitment in your organization or place of work, or management at your
job or internship position.” The research was approved by the institutional review
board of the university and survey data were collected electronically in spring 2009
through a Qualtrics online data gathering system hosted on the university servers.
4.1 Measures
The internal factor structure of the three measures was examined using factor
analyses employing principal components analysis (PCA) with Varimax rotation as
described below.
4.1. i. Trust in top management (organizational trust) and trust in immediate
supervisor (managerial trust) – Ellis and Shockley-Zalabak (2001). Ellis and
Shockley-Zalabak’s (2001) behavior-based measure assessed the extent to which top
management demonstrates behaviors that exemplify trust. As can be seen from
Table I, examining the underlying dimensionality of the scale by factor analyses
employing principal components with Varimax rotation obtained one factor with
eigenvalues 41.0 explaining 67.24 percent of the variance (eigenvalue ¼4.03; Table I).
The scale reliability with six items in the present study was high (a¼0.90 for items in
trust in top management, M¼3.56, SD ¼0.80, range: 3.41-3.81).
4.1. ii. Organizational trust – Hon and Grunig (1999). Factor analyses employing
PCA with Varimax rotation revealed three underlying dimensions with eigenvalues
41 and explaining a total variance of 69.14 percent. The organizational trust scale
employed by Hon and Grunig (1999) consisted of the three dimensions of integrity,
dependability, and competence. Two items were dropped from this scale: “I think it
is important to watch this organization closely so that it does not take advantage of
people like me” (recoded) – loaded on factor 3; and “This organization is known to be
successful at the things it tries to do” – because it cross-loaded on factors 1 and 2.
F. 1
Eigenvalues/% variance explained M(SD) Range a4.03/67.24
I. Trust in top management 3.56 (0.80) 3.41-3.81 0.90
I trust top management 3.81 (0.94) 1.00-5.00 0.78
Top management is sincere in their efforts to
communicate with employees 3.65 (1.01) 1.00-5.00 0.81
Top management listens to employees’ concerns 3.43 (0.10) 1.00-5.00 0.85
Top management keeps its commitments to employees 3.51 (0.98) 1.00-5.00 0.82
Top management is concerned about employees’
well being 3.57 (0.10) 1.00-5.00 0.78
Those in top management keep their word to employees 3.41 (0.94) 1.00-5.00 0.86
Source: Ellis and Shockley-Zalabak (2001)
Tabl e I.
Factor structure and
internal reliability for
trust in top management
30
JCOM
17,1
The internal reliability of the remaining nine items was good (a¼0.89, M¼3.57,
SD ¼0.89, range: 3.28-3.93) (Table II).
4.1. iii. Interpersonal trust in top management – Cook and Wall (1980). The Cook
and Wall (1980) scale develops instruments relevant to multivariate research into the
quality of working life and conceptualizes trust as the extent to which one is willing to
ascribe good intentions to and have confidence in the words and actions of other
people. This scale examines trust along two different dimensions: faith in trustworthy
intentions of others, and confidence in ability of others (management). As Table III
shows, the internal reliability of trust in management is good (a¼0.84, M¼3.64,
SD ¼0.71, range ¼3.54-3.74) and with two factors with eigenvalues 41 explaining
a total variance of 71.13 percent (Table III). Upon conducting exploratory factor
analyses, all items with factor loadings above 0.70 were retained and loaded on
factor 1.One item loaded on factor 2 and was dropped (“our management would
F. 1 F. 2 F. 3
Eigenvalues/% of variance explained M(SD) Range a5.41/49.25 1.14/10.39 1.05/9.55
Organizational trust 3.52 (0.63) 3.28-3.93 0.89
This organization treats people like me
fairly and justly 3.86 (0.83) 1.00-5.00 0.72 0.23 0.17
Whenever this organization makes an
important decision, I know it will be
concerned about people like me 3.29 (1.03) 1.00-5.00 0.69 0.35 0.12
This organization can be relied upon to
keep its promises 3.61 (0.82) 1.00-5.00 0.80 0.15 0.06
I believe that this organization takes
the opinions of people like me into
account when making decisions 3.33 (0.97) 1.00-5.00 0.68 0.47 0.30
I feel very confident about this
organization’s skills 3.65 (0.83) 1.00-5.00 0.81 0.14 0.02
This organization has the ability to
accomplish what it says it will do 3.94 (0.73) 1.00-5.00 0.63 0.45 0.43
Sound principles seem to guide this
organization’s behavior 3.57 (0.87) 1.00-5.00 0.73 0.21 0.08
This organization does not mislead
people like me 3.62 (0.90) 1.00-5.00 0.81 0.01 035
I am very willing to let this
organization make decisions for people
like me 3.28 (1.00) 1.00-5.00 0.67 0.02 016
I think it is important to watch this
organization closely so that it does not
take advantage of people like me* 3.13 (1.14) 1.00-5.00 0.47 0.32 0.69
This organization is known to be
successful at the things it tries to do 3.94(0.83) 1.00-5.00 0.61 0.59*0.31
Notes: The italicized items were italicized to denote that they were deleted from the scale because they
loaded poorly on the scale as denoted by the corresponding italicized values reported under column F. 1,
F. 2, and F. 3. Individual alpha values are not reported for each item while reporting alpha reliability of
the overall scale. The alpha reliability of the scale comprising these individual items is reported under
the “Organizational Trust” (in bold) at the top of the scale. Hence this is appropriate as it stands in the
table (see also the “Measures” section in the paper)
Source: Hon and Grunig Scale (1999)
Table II.
Factor structure and
internal reliability for
organizational trust
31
Organizational
trust
be quite prepared to gain advantage by deceiving the members”; recoded, loaded on
factor 2; Table III).
4.1. iv. Organizational identification – Cheney (1983). Exploratory factor analysis
with principal components with Varimax rotation obtained five factors with
eigenvalues 41 and explaining a total of 64.83 percent of the variance. Examining
the underlying dimensionality of the items, all of the items loaded on factor 1 with the
exception of the following items which were dropped from the final measure because
of cross-loading on multiple dimensions or loading on a different factor: “I often
describe myself to others” by saying “I work for the organization” or “I am from the
organization”; “I try to make on-the-job decisions by considering the consequences of
my actions for the organization”; “I find it difficult to agree with the organization’s
policies relating to me” (recoded); “My association with the organization is only a small
part of who I am” (recoded); “I feel very little loyalty to the organization” (recoded). The
internal reliability of the final measure comprising 20 items was excellent at a¼0.94
(M¼3.281, SD ¼1.01, range ¼2.23-3.62; scale from Rubin et al., 1994).
5. Convergent validity analysis
Campbell and Fiske (1959) propose the MTMM matrix offering a set of criteria for
validating measurements including: confirmation of validation of the measurements
by independent measurement procedures; conducting discriminant validation as well
as convergent validation to establish high correlations with similar tests and low
correlations with those with which the measures are intended to be conceptually
F. 1 F. 2
Eigenvalues/% variance explained M(SD) Range a3.21/53.45 1.06/17.68
I. Trust in top management faith in
intentions of management 3.64 (0.71) 3.54-3.74 0.84
Management at my firm is sincere in its
attempts to meet the members’ point of view 3.54 (0.88) 1.00-5.00 0.71 0.26
I feel quite confident that the organization will
always try to treat me fairly 3.68 (0.91) 1.00-5.00 0.76 0.269
Our management would be quite prepared to
gain advantage by deceiving the members* 3.65 (1.16) 1.00-5.00 0.35 0.87
Confidence in actions of management
Our organization has a poor future unless it
can attract better managers 3.56 (1.12) 1.00-5.00 0.77 0.33
Management can be trusted to make sensible
decisions for the organization’s future 3.74 (0.78) 1.00-5.00 0.82 0.02
Management at work seems to do an efficient
job 3.67 (0.84) 1.00-5.00 0.86 0.19
Notes: The main scale in bold (“Trust in Top Management”) reports the alpha reliability of the scale
(0.84). The other two bold statements denoting the dimensions were not treated separately but as part
of the overall scale (“Trust in Top Management”) and hence there is no individual alpha reliability to
report for these two bold statements denoting dimensions (“Faith in intentions of management” and
“Confidence in actions of management”). The items from these individual dimensions were treated
together as part of the overall construct “Trust in Top Management” and therefore the alpha reliability
of this measure is reported (0.84)
Source: Cook and Wall Scale (1980)
Table III.
Factor structure and
internal reliability for
trust in top management
32
JCOM
17,1
distinct, conducting tests as a trait-method unit or “a union of a particular trait
content with measurement procedures not specific to that content” (Campbell and
Fiske, 1959, p. 81); and, employing more than one trait and more than one method in the
discriminant validation process.
The MTMM matrix essentially measures all the correlations of the different traits
when measured with different methods. Convergent validity is established when
measures that are conceptually correlated are demonstrated to be so (as e.g.
organizational trust and organizational identification). Per the MTMM matrix, in order
to establish convergent validity, the validity diagonal should meet two criteria: it
should be large and different from zero, and, it should be higher than the correlations
obtained between that variable having neither the trait nor the method in common.
In other words, if the measure is assessed with respect to a similar construct
(within construct correlations) and with respect to a conceptually distinct construct
(cross-construct correlations), the within construct correlations should be high and
the cross-construct correlations should be low. In accordance with the MTMM
(Campbell and Fiske, 1959) matrix, convergent validity of the organizational trust scale
was assessed following the principle that measures of theoretically similar constructs
should be highly intercorrelated. In order to assess convergent validity: the strength of
inter-item correlations within scale were examined, and cross-construct correlations
across trust in top management trust scales and a theoretically correlated construct,
identification, was assessed.
Preliminary analyses examining the dimensionality of the organizational trust
scales using PCA in order to examine the underlying factor structure and validity were
conducted prior to examining the MTMM matrix. Exploratory factor analyses
were conducted using PCAs with unrotated factor solution and Varimax rotation.
Factors with eigenvalues 41.0 were retained and primary factor loadings 40.65 with
secondary loadings no 40.45. Items that did not load satisfactorily or cross-loaded
were dropped. Items that were retained in the final measures are marked with boldface
and those that were dropped are italicized (Tables I, II, and III). CFAs (Bollen, 1989;
Kline, 2004) were conducted with the remaining items to assess factor structure and
validity with the retained items. The first set of CFAs assessed a measurement model
comprising the unidimensional Hon and Grunig (1999) organizational trust scale, the
Ellis and Shockley-Zalabak (2001) scale and the Cook and Wall (1980) scale. The second
CFA assessed model fit of the trust and identification measure.
6. Results
The hypotheses proposed predicted that if the Hon and Grunig (1999) organizational
trust measure from the public relations literature is a valid construct, we would expect
to see: a positive associational relationship between organizational trust scores with
the organizational communication literature (H1). Further, we would also expect to see
positive correlations with the organizational identification scale similar in strength
and direction with those obtained from correlations with the other organizational
communication trust scales (H2a; Cook and Wall, 1980; Ellis and Shockley-Zalabak,
2001), and the internal reliability of the Hon and Grunig (1999) scale will be comparable
to that of the other two measures of trust in top management (H2b).
H1 was tested through Pearson moment correlations with the trust (Hon and
Grunig, 1999) scale and the trust in top management (Ellis and Shockley-Zalabak,
2001) and the trust in management (Cook and Wall, 1980) scale. The results provide
strong support for this assertion: the correlation between trust scale (Hon and Grunig,
33
Organizational
trust
1999) and Ellis and Shockley-Zalabak (2001) were strong, positive, and significant
(r¼0.82, po0.001, Table IV). Similarly theoretically, we would expect a positive
relationship between subject scores on the Hon and Grunig (1999) trust scale and
the ratings of Cook and Wall (1980) trust scale. Again, results provide strong
support for this assertion as well (r¼0.80, po0.001). As the goodness-of-fit indices
reported in Table V show, the measurement model CFA indicated a mixed fit for the
correlated Hon and Grunig, Ellis and Shockley-Zalabak, and Cook and Wall scale.
While modification indices achieved an excellent model fit to the observed data, upon
testing alternative measurement models correlating the Hon and Grunig scale with
the Cook and Wall scale and with the Ellis and Shockley-Zalabak scale, a mixed fit
was obtained again (Table V).
H2a was tested through Pearson moment correlations with the Hon and Grunig
measure and Cheney’s organizational identification measure. As expected, the
association was strong, positive, and significant (r¼0.66, po0.001; Table IV).
This result was in congruence with the associations between the Ellis and Shockley-
Zalabak (2001) trust in top management scale and Cheney’s organizational
identification scale (r¼0.53, po0.001) as well as with the Cook and Wall (1980)
trust in management scale and Cheney’s organizational identification scale (r¼0.70,
Model w
2
(df), pw
2
/df RMSEA
H1a Measurement model for correlated Hon and
Grunig, Ellis and Shockley-Zalabak, and Cook
and Wall scale
w
2
(167) ¼309.09, po0.001 1.85 0.09
With modification indices w
2
(151) ¼177.17,po0.10 1.17 0.04
Alternative model correlating Hon and
Grunig with the Cook and Wall scale
w
2
(76) ¼151.68, po0.001 2.00 0.89
-alternative measurement model for
correlated Hon and Grunig and Ellis and
Shockley-Zalabak scale
w
2
(89) ¼143.56, po0.001 1.61 0.07
H2a Measurement model for correlated Hon and
Grunig, Ellis and Shockley-Zalabak, and Cook
and Wall scale
w
2
(734) ¼1,286.89, po0.000 1.75 0.08
Alternative model combining the
organizational identification and Ellis and
Shockley-Zalabak (2001) scale correlated with
Hon and Grunig scale
w
2
(559) ¼1,164.82, po0.000 2.08 0.10
Note: n¼108
Tabl e V.
Goodness-of-fit statistics
for measurement models
and alternative models
testing H1a and H2a
Scale 1 2 3 4
Organizational trust (Ellis and Shockley-Zalabak, 2001) 0.94 0.82
*
0.59
*
0.53
*
Organizational trust (Hon and Grunig, 1999) 0.82
*
0.89 0.80
*
0.66
*
Trust in top management (Cook and Wall, 1980) 0.59
*
0.80
*
0.84 0.70
*
Organizational identification (Cheney, 1983) 0.53
*
0.66
*
0.70
*
0.94
Notes: Items in bold are alpha reliabilities; items in italics are cross construct correlations. *po0.001
Tabl e IV.
Correlations matrix for
trust in top management
and identification
measurement
34
JCOM
17,1
po0.001; Table IV). However, the CFA measurement model indicated a marginally
adequate fit for the model correlating all three trust measures (see Table V for
goodness-of-fit indices). Following Ellis and Shockley-Zalabak’s study finding close
conceptual association of organizational identification and organizational trust, an
alternative model was tested combining organizational identification and Ellis and
Shockley-Zalabak measures and correlated with Hon and Grunig’s measure which
resulted in a decreased model fit (Table V) suggesting that specifying organizational
identification as a separate latent variable had a marginally better fit to the data.
As can be seen from Table IV, H2b was supported. The internal reliability after
revising on the basis of the exploratory factor analyses of the Hon and Grunig
(1999) organizational trust scale was satisfactory (a¼0.89; n¼108) and compared
favorably to the other two measures of trust in top management: the Cook and Wall
organizational trust awas 0.84 (n¼108), and the Ellis and Shockley-Zalabak’s awas
0.90 (n¼108) indicating excellent internal consistency (Nunnally, 1978). Scores on
the organizational trust measure from Hon and Grunig (1999) ranged from 3.28 to 3.93
with a mean score of 3.52 (SD ¼0.63). This score would appear to be comparable
to a mean score of 3.56 (SD ¼0.80) on the Ellis and Shockley-Zalabak (2001) trust in
top management scale and 3.64 (SD ¼0.71) on the Cook and Wall (1980) trust in top
management scale. The scales have an expected range and appear to be able to capture
individual variability well for an undergraduate student sample.
7. Discussion
The present study is among the first to assess the construct validity of organizational
trust measures from three perspectives: organizational communication, public
relations, and interpersonal communication at work. The study findings provide
mixed evidence for the convergent validity of the three measures of organizational
trust. Convergent validity was tested through theoretically based within construct and
cross-construct correlations. Strong Pearson moment correlations between the three
organizational trust self-report scales suggest that the three measurements are highly
associated. While the marginally adequate fit of the measurement model indicates that
the error variances of the three trust measures might be correlated.
Second, Pearson correlations indicate the three trust measures also correlate
significantly and positively with a related constr uct, identification. Again, CFAs of
the measurement model demonstrated mixed goodness-of-fit statistics ranging from
marginally adequate to acceptable to the data. The test of an alternative model with the
identification indicators modeled on the Ellis and Shockley-Zalabak (2001) trust scale
and correlated with Hon and Grunig’s (1999) scale did not result in an improved model
fit. In support of the hypothesis, all three trust measures rely upon an understanding
of trust as an individual behavior-based orientation and demonstrate comparable
and excellent internal consistency and homogeneity of variances as seen through the
reliabilities obtained in the study. Further, the strength of the Pearson moment
correlations supports the assertion that the measures are conceptually reflective of
same theoretical base and provide a degree of confidence in the convergent validity
of the measure.
In other words, as hypothesized, although the measures take different approaches
(ranging from interpersonal trust at work to a public relations perspective), the
underlying assumptions of trust are measuring the same dimensions of organizational
trust. In this context, one possibility explaining the mediocre model fit obtained by the
CFAs could be the correlated error variances as demonstrated by the modification
35
Organizational
trust
indices needed to obtain good model fit of the CFA measurement model. A potential
cause of the measurement error in the three measures of trust could be methods
bias caused by study participants responding consistently higher or lower on the
self-reports. This could be due to participants’ perception of social desirability of
the response or an underlying redundancy of the items (Kline, 2004). The study results
suggest that the assessment of organizational trust measures through only self-reports
runs a high risk of measurement error due to systematic methods bias.
While the strong associations between the three measures suggest they are
measuring the same underlying concept, the study findings suggest the need for
greater conceptual clarification of the theoretical bases of trust. The correlations
between trust and identification assume that all measurement error terms are zero.
As the study demonstrates, both measurement models correlating identification with
the three trust measures and modeling identification as part of organizational trust
(Ellis and Shockley-Zalabak’s measure) and correlated with Hon and Grunig’s (1999)
measure were tested and provided good support for the convergent validity of
the organizational trust measures. Cheney (1983) suggests that organizational
identification is an essential part of the process whereby members adopt the
underlying value premises and decision premises of the organization. Ellis and
Shockley-Zalabak (2000) have examined identification as a part of trust-building
process while others (e.g. Cook and Wall, 1980; Hon and Grunig, 1999) treat
identification as strongly related but conceptually distinct from organizational trust.
The findings suggest that the conceptual ground between identification and trust
should be rigorously explored to understand the shared conditions of their evolution,
manifestation, and decline.
In conclusion, the study finds that while the three organizational trust measures
are conceptually associated, as demonstrated by the correlational strengths
indicating substantial convergent validity, the results suggest steps for
strengthening the convergent validity of trust including: increasing item parsimony
to decrease redundancy, clarifying item construction to account for correlated error
variance due to factors such as social desirability of response, and improving
theoretical specification investigating whether organizational identification is a factor
of organizational trust or an antecedent or outcome variable.
7.1 Limitations
One limitation of the study is the correlational matrices rely on self-reports and do not
investigate method variance. Second, the study participants provide a part-time and
temporary work experience. Within these limitations, the results of the present study
provide an important first step toward an integrative assessment of this important
phenomenon in communication literature. Future studies can extend the present study
by examining common method and trait variance as well as by investigating the
construct validity of organizational trust with members from different organizational
levels, industries, and regions.
7.2 Significance and future research
The study is the first to demonstrate the convergent validity of organizational
trust measures and contributes to existing literature by taking a first step toward
establishing that the measures employed to investigate trust in scholarship are indeed,
measuring the construct they intend to examine. Future research can extend the results
of the present study by conducting tests of the discriminant validity of organizational
36
JCOM
17,1
trust measures. Future research should clarify the role of organizational identification
as a correlate, outcome, or factor of trust to understand their relationship with trust
antecedents (exchange-based, knowledge-based, or identification-based).
References
Argyris, C. (1962), Interpersonal Competence and Organizational Effectiveness,Dorsey,
Homewood, IL.
Bhattacharya, R., Devinney, T.M. and Pillutla, M.M. (1998), “A formal model of trust based on
outcomes”, Academy of Management Journal, Vol. 23 No. 3, pp. 459-72.
Bollen,K.A.(1989),Structural Equations with Latent Variables, John Wiley & Sons, New York, NY.
Bullis, C. and Bach, B.W. (1991), “An explication and test of communication network content and
multiplexity as predictors of organizational identification”, Western Journal of Speech
Communication, Vol. 55 No. 2, pp. 180-97.
Burke, K. (1950), A Rhetoric of Motives, University of California Press, Berkeley, CA.
Campbell, D.T. and Fiske, D.W. (1959), “Convergent and discriminant validity by the multitrait
multimethod matrix”, Psychological Bulletin, Vol. 56 No. 2, pp. 81-105.
Cheney, G. (1983), “The rhetoric of identification and the study of organizational communication”,
Quarterly Journal of Speech, Vol. 69 No. 2, pp. 143-58.
Cheney, G. and Tompkins, P.K. (1987), “Coming to terms with organizational identification and
commitment”, Central States Speech Journal, Vol. 38 No. 1, pp. 1-15.
Coleman, J.S. (1990), Foundations of Social Theory, Harvard University Press, Cambridge, MA.
Cook, J. and Wall, T. (1980), “New work attitude measures of trust, organizational commitment,
and personal need non-fulfillment”, Journal of Occupational Psychology, Vol. 53 No. 1,
pp. 39-52.
Creed, W.E.D. and Miles, R.E. (1996), “Trust in organizations: a conceptual framework linking
organizational forms, managerial philosophies, and the opportunity cost of controls”, in
Kramer, R.M. and Tyler, T.R. (Eds), Trust in Organizations: Frontiers of Theory and
Research, Sage Publications Inc, Thousand Oaks, CA, pp. 16-38.
Cronbach, L.J. and Meehl, P.E. (1955), “Construct validity in psychological tests”, Psychological
Bulletin, Vol. 52 No. 4, pp. 281-302.
Cummings, L.L. and Bromiley, P. (1996), “The organizational trust inventory (OTI): development
and validation”, in Kramer, R.M. and Tyler, T.R. (Eds), Trust in Organizations: Frontiers of
Theory and Research, Sage Publications Inc, Thousand Oaks, CA, pp. 302-30.
Das, T.K. and Teng, B. (1998), “Between trust and control: developing confidence in partner
cooperation in alliances”, Academy of Management Review, Vol. 23 No. 3, pp. 491-512.
Dasgupta, P. (1998), “Trust as a commodity”, in Gambetta, D.G. (Ed.) Trust, Basil Blackwell,
New York, NY, pp. 49-72.
Ellis, K. and Shockley-Zalabak, P. (2001), “Trust in top management and immediate supervisor:
the relationship to satisfaction, perceived organizational effectiveness, and information
receiving”, Communication Quarterly, Vol. 49 No. 4, pp. 382-99.
Gabarro, J. (1987), The Dynamics of Taking Charge, Harvard Business School Press, Boston, MA.
Hardin, R. (2000), Trust and Trustworthiness, Russell Sage Foundation, New York, NY.
Hon, L.C. and Grunig, J.E. (1999), Guidelines for Measuring Relationships in Public Relations,
Institute for Public Relations, Gainesville, FL.
Jarvenpaa, S., Knoll, K. and Leidner, D. (1998), “Is anybody out there?: the development
of trust in virtual teams”, Journal of Management Information Systems, Vol. 14 No. 4,
pp. 29-64.
37
Organizational
trust
Jones, G.R. and George, J.M. (1998), “The experience and evolution of trust: implications
for cooperation and teamwork”, The Academy of Management Review, Vol. 23 No. 3,
pp. 531-46.
Kline, R.B. (2004), Principles and Practice of Structural Equation Modeling, 2nd ed., The Guilford
Press, New York, NY.
Larzelere, R. and Huston, T. (1980), “The dyadic trust scale: toward understanding interpersonal
trust in close relationships”, Journal of Marriage and the Family, Vol. 42 No. 3,
pp. 595-604.
Lee, S.M. (1971), “An empirical analysis of organizational identification”, Academy of
Management Journal, Vol. 14 No. 2, pp. 213-26.
Lewicki, R. and Bunker, B. (1995), “Developing and maintaining trust in work relationships”, in
Kramer, R. and Tyler, T. (Eds), Trust in Organizations: Frontiers in Theory and Research,
Sage, Newbury Park, CA, pp. 114-39.
McAllister, D. (1995), “Affect-and cognition-based trust as foundations for interpersonal
cooperation in organizations”, Academy of Management Journal, Vol. 38 No. 1, pp. 24-59.
McGregor, D. (1967), The Professional Manager, McGraw-Hill, New York, NY.
March, J.G. and Simon, H.A. (1958), Organizations, John Wiley & Sons, New York, NY.
Mayer, R.C., Davis, J.H. and Schoorman, F.D. (1995), “An integrative model of organizational
trust”, Academy of Management Review, Vol. 20 No. 3, pp. 709-34.
Mishra, A.K. (1996), “Organizational response to crisis: the centrality of trust”, in Kramer, R.M.
and Tyler, T.R. (Eds), Trust in Organizations: Frontiers of Theory and Research, Sage,
Thousand Oaks, CA, pp. 261-87.
Nunnally, J.C. (1978), Psychometric Theory, 2nd ed., McGraw-Hill, New York, NY.
Rotter, J.B. (1967), “A new scale for the measurement of interpersonal trust”, Journal of
Personality, Vol. 35 No. 4, pp. 651-65.
Rotter, J.B. (1971), “Generalized expectancies for interpersonal trust”, American Psychologist,
Vol. 26 No. 5, pp. 443-52.
Rotter, J.B. (1980), “Interpersonal trust, trustworthiness, and gullibility”, American Psychologist,
Vol. 35 No. 1, pp. 1-7.
Rousseau, D.M. (1990), “Assessing organizational culture: the case for multiple methods”, in
Schneider, B. (Ed.), Organizational Climate and Culture, Jossey-Bass, San Francisco, CA,
pp. 153-92.
Rousseau, D.M., Sitkin, S., Burt, R. and Camerer, C. (1998), “Not so different after all: a cross
disciplinary view of trust”, Academy of Management Review, Vol. 23 No. 3, pp. 387-92.
Rubin, R.B., Palmgreen, P. and Sypher, H.E. (1994), Communication Research Measures: A
Sourcebook, The Guilford Press, New York, NY.
Schein, E.H. (1996), “Culture: the missing concept in organization studies”, Administrative
Science Quarterly, Vol. 41 No. 2, pp. 229-40.
Shockley-Zalabak, P., Ellis, K. and Winngrad, G. (2000), “Organizational trust: what it means,
why it matters”, Organizational Development Journal, Vol. 18 No. 4, pp. 35-47.
Tannenbaum, S.I., Beard, R.L. and Salas, E. (1992), “Team building and its influence on team
effectiveness: an examination of conceptual and empirical evidence”, in Kelley, K. (Ed.),
Issues, Theory and Research in Industrial/Organizational Psychology, Elsevier Science
Publishers, Amsterdam, pp. 503-29.
Tompkins, P.K. and Cheney, G. (1983), “Account analysis in organizations: decision making and
identification”, in Putnam, L.L. and Pacanowsky, M.E. (Eds), Communication and
Organizations: An Interpretive Approach, Sage, Beverly Hills, CA, pp. 123-46.
38
JCOM
17,1
Whitener, E.M., Brodt, S.E., Korsgaard, M.A. and Werner, J.M. (1998), “Managers as initiators of
trust: an exchange relationship framework for understanding managerial trustworthy
behavior”, Academy of Management Review, Vol. 23 No. 3, pp. 513-30.
Zucker, L.G. (1986), “Production of trust: institutional sources of economic structure, 1840-1920”,
Research in Organizational Behavior, Vol. 8, JAI, Greenwich, CT, pp. 53-111.
Further reading
Eid, M., Nussbeck, F.W., Geiser, C., Cole, D.A., Gollwitzer, M. and Lischetzke, T. (2008),
“Structural equation modeling of multitrait-multimethod data: different models for
different types of methods”, Psychological Methods, Vol. 13 No. 3, pp. 230-53.
Hubbell, A.P. and Chory-Assad, R.M. (2005), “Motivating factors: perceptions of justice and their
relationship with managerial and organizational trust”, Communication Studies, Vol. 56
No. 1, pp. 47-70.
Hubbell, A.P. and Medved, C.E. (2001), “Measuring trustworthy organizational behaviors: a
review of the literature and scale validation”, paper presented at the Organizational
Communication Division of the Annual Meeting of the International Communication
Association, May 24-28, Washington, DC.
Korsgaard, M.A., Brodt, S.E. and Whitener, E.M. (2002), “Trust in the face of conflict: the role
of managerial trustworthy behavior and organizational context”, Journal of Applied
Psychology, Vol. 87 No. 2, pp. 312-9.
Westen, D. and Rosenthal, R. (2005), “Improving construct validity: Cronback, Meehl, and
Neurath’s ship”, Psychological Assessment, Vol. 17 No. 4, pp. 409-12.
About the author
Vinita Agarwal, PhD, is an Assistant Professor of Communication at the Salisbury University.
She earned her Doctorate in Communication from Purdue University in May, 2009. Vinita
Agarwal can be contacted at: vxagarwal@salisbury.edu
To purchase reprints of this article please e-mail: reprints@emeraldinsight.com
Or visit our web site for further details: www.emeraldinsight.com/reprints
39
Organizational
trust
Reproducedwithpermissionof thecopyrightowner.Furtherreproduction prohibitedwithoutpermission.
